Files
enginex-hygon-llama.cpp/benches/dgx-spark/run-aime-120b-t8-x8-high.log

12 lines
17 KiB
Plaintext
Raw Normal View History

nohup: ignoring input
Running with args Namespace(model='openai/gpt-oss-120b', reasoning_effort='high', sampler='chat_completions', base_url='http://localhost:8066/v1', eval='aime25', temperature=1.0, n_threads=8, debug=False, examples=None)
Running the following evals: {'aime25': <gpt_oss.evals.aime_eval.AIME25Eval object at 0xe8638c0857c0>}
Running evals for the following models: {'openai/gpt-oss-120b-high': <gpt_oss.evals.chat_completions_sampler.ChatCompletionsSampler object at 0xe8638c085340>}
0%| | 0/240 [00:00<?, ?it/s] 0%| | 1/240 [02:55<11:38:24, 175.33s/it] 1%| | 2/240 [03:48<6:51:24, 103.72s/it] 1%|▏ | 3/240 [05:47<7:16:38, 110.54s/it] 2%|▏ | 4/240 [06:44<5:51:23, 89.34s/it] 2%|▏ | 5/240 [07:53<5:21:29, 82.08s/it] 2%|▎ | 6/240 [17:13<15:53:25, 244.47s/it] 3%|▎ | 7/240 [25:07<20:41:12, 319.62s/it] 3%|▎ | 8/240 [25:10<14:06:28, 218.91s/it] 4%|▍ | 9/240 [27:13<12:06:34, 188.72s/it] 4%|▍ | 10/240 [30:33<12:16:51, 192.22s/it] 5%|▍ | 11/240 [32:45<11:03:23, 173.82s/it] 5%|▌ | 12/240 [39:34<15:31:54, 245.24s/it] 5%|▌ | 13/240 [44:17<16:11:28, 256.78s/it] 6%|▌ | 14/240 [51:54<19:55:10, 317.30s/it] 6%|▋ | 15/240 [52:26<14:26:56, 231.18s/it] 7%|▋ | 16/240 [1:00:47<19:26:26, 312.44s/it] 7%|▋ | 17/240 [1:00:52<13:37:59, 220.09s/it] 8%|▊ | 18/240 [1:09:47<19:24:30, 314.73s/it] 8%|▊ | 19/240 [1:15:59<20:22:08, 331.80s/it] 8%|▊ | 20/240 [1:19:18<17:51:04, 292.11s/it] 9%|▉ | 21/240 [1:23:46<17:19:23, 284.77s/it] 9%|▉ | 22/240 [1:27:49<16:29:35, 272.36s/it] 10%|▉ | 23/240 [1:31:20<15:18:04, 253.85s/it] 10%|█ | 24/240 [1:33:42<13:12:51, 220.24s/it] 10%|█ | 25/240 [1:34:30<10:03:44, 168.49s/it] 11%|█ | 26/240 [1:35:12<7:45:29, 130.51s/it] 11%|█▏ | 27/240 [1:41:05<11:40:47, 197.40s/it] 12%|█▏ | 28/240 [1:41:52<8:58:11, 152.32s/it] 12%|█▏ | 29/240 [1:50:38<15:29:30, 264.31s/it] 12%|█▎ | 30/240 [1:58:51<19:25:22, 332.97s/it] 13%|█▎ | 31/240 [2:06:54<21:57:01, 378.09s/it] 13%|█▎ | 32/240 [2:11:15<19:49:00, 342.98s/it] 14%|█▍ | 33/240 [2:15:26<18:07:41, 315.27s/it] 14%|█▍ | 34/240 [2:28:06<25:40:44, 448.76s/it] 15%|█▍ | 35/240 [2:36:39<26:38:46, 467.93s/it] 15%|█▌ | 36/240 [2:50:14<32:24:40, 571.96s/it] 15%|█▌ | 37/240 [2:58:42<31:10:55, 552.98s/it] 16%|█▌ | 38/240 [3:02:52<25:55:32, 462.04s/it] 16%|█▋ | 39/240 [3:07:01<22:13:39, 398.11s/it] 17%|█▋ | 40/240 [3:08:19<16:47:08, 302.14s/it] 17%|█▋ | 41/240 [3:17:10<20:29:05, 370.58s/it] 18%|█▊ | 42/240 [3:17:43<14:49:16, 269.48s/it] 18%|█▊ | 43/240 [3:25:41<18:10:02, 331.99s/it] 18%|█▊ | 44/240 [3:37:25<24:08:45, 443.50s/it] 19%|█▉ | 45/240 [3:47:01<26:11:16, 483.47s/it] 19%|█▉ | 46/240 [3:48:23<19:33:53, 363.06s/it] 20%|█▉ | 47/240 [3:53:08<18:12:17, 339.57s/it] 20%|██ | 48/240 [4:10:36<29:26:13, 551.95s/it] 20%|██ | 49/240 [4:10:53<20:46:01, 391.42s/it] 21%|██ | 50/240 [4:17:30<20:45:27, 393.30s/it] 21%|██▏ | 51/240 [4:29:11<25:29:12, 485.46s/it] 22%|██▏ | 52/240 [4:34:22<22:36:56, 433.07s/it] 22%|██▏ | 53/240 [4:47:51<28:21:08, 545.82s/it] 22%|██▎ | 54/240 [4:52:15<23:50:38, 461.50s/it] 23%|██▎ | 55/240 [4:54:52<19:00:49, 370.00s/it] 23%|██▎ | 56/240 [4:55:15<13:35:13, 265.84s/it] 24%|██▍ | 57/240 [5:00:00<13:48:46, 271.73s/it] 24%|██▍ | 58/240 [5:00:48<10:20:53, 204.69s/it] 25%|██▍ | 59/240 [5:03:49<9:55:38, 197.45s/it] 25%|██▌ | 60/240 [5:06:15<9:06:35, 182.20s/it] 25%|██▌ | 61/240 [5:09:56<9:37:40, 193.64s/it] 26%|██▌ | 62/240 [5:13:47<10:07:58, 204.94s/it] 26%|██▋ | 63/240 [5:14:29<7:40:38, 156.15s/it] 27%|██▋ | 64/240 [5:17:05<7:37:56, 156.11s/it] 27%|██▋ | 65/240 [5:30:00<16:36:50, 341.78s/it] 28%|██▊ | 66/240 [5:39:18<19:39:03, 406.57s/it] 28%|██▊ | 67/240 [5:42:53<16:46:13, 348.98s/it] 28%|██▊ | 68/240 [5:48:27<16:27:58, 344.64s/it] 29%|██▉ | 69/240 [5:55:10<17:11:40, 361.99s/it] 29%|██▉ | 70/240 [6:00:11<16:14:13
Writing report to /tmp/aime25_openai__gpt-oss-120b-high_temp1.0_20251109_094547.html
{'chars': np.float64(2296.1916666666666), 'chars:std': np.float64(986.051306946325), 'score': np.float64(0.925), 'score:std': np.float64(0.26339134382131846)}
Writing results to /tmp/aime25_openai__gpt-oss-120b-high_temp1.0_20251109_094547.json
Writing all results to /tmp/aime25_openai__gpt-oss-120b-high_temp1.0_20251109_094547_allresults.json
[{'eval_name': 'aime25', 'model_name': 'openai__gpt-oss-120b-high_temp1.0_20251109_094547', 'metric': 0.925}]